An Energy-Efficient Secure Adaptive Cloud-of-Things (CoT) Framework to Facilitate Undergraduate STEM Education.

This paper presents an adaptive framework (web application) that supports secure, end-to-end Internet of Things (IoT) sensing architecture, which includes loosely coupled set of components with well-defined interfaces that can be adapted to a wide range of uses. Moreover, the framework was adopted in a Computer Science (CS) Fog computing course to get formal evaluation by undergraduate students. IoT has emerged as a new network paradigm, involving connecting physical world through sensors (temperature, humidity, intensity, etc.) to the Internet, to provide opportunities for building intelligent systems and applications. There are a few notable characteristics of today's IoT systems that serve as impediments for greater adoption. First, the common, well-documented platform that spans end-to-end, and which can cover a multitude of practical uses, is still missing. The second impediment is the lack of standard energy-efficient platform. Third, with the growing deployment of IoT devices, the attack vector for possible exploitation increases, which dictates an increase in the ongoing cost to ensure Confidentiality-Integrity-Availability triad. In this project, we propose the solution to address these impediments. The proposed framework makes it economical and easy to adapt to (e.g. to learn and to support). The next benefit comes with the use of environment harvesting equipment (solar cell, in-line hydro-generators, thermoelectric cells, etc.) and use of energy efficient coding methods, including security and communication provisions. Overall, the project successfully integrates Amazon Cloud technologies, Raspberry-Pi and Arduino embedded systems using energy-harvesting technologies, secure communication using government standards, and highly available network of nodes using wireless technology into a Cloud-of-Things (CoF).The methodology allows a myriad IoT devices connected to the Internet to be easily monitored and manipulated through the implemented web application. On the other hand, since few dedicated IoT courses are currently offered, most Science, technology, Engineering, and Mathematics (STEM) students will have limited or no exposure to IoT development until after graduation, and inadequately prepared in IoT development to enter into the workforce. Therefore, the implemented framework was introduced in a CS special topic course to get comments and formal evaluations from undergraduate STEM students. An outline of a well-developed courseware is presented in the paper. The courseware is evaluated through survey questions. The majority of the students provided positive feedback and enjoyed the IoT framework.